用相似性记忆引导大模型推理，提升长链条任务准确率
Directional Attractors in LLM Reasoning: How Similarity Retrieval Steers Iterative Summarization Based Reasoning
- 引入语义缓存，检索过往成功推理模式指导当前推理
- 在MATH500等数据集上准确率提升，但跨领域任务出现退化
- 揭示了记忆检索带来的推理方向性偏差，适合研究自进化模型
基于迭代摘要的推理框架如InftyThink通过控制上下文增长实现大语言模型的长程推理，但会重复生成相似推理策略。我们提出InftyThink with Cross-Chain Memory，引入基于嵌入的语义缓存，存储过往成功的推理片段。每一步推理中，模型检索并条件于最相似的已存储命题，引导推断而不盲目扩展上下文。在MATH500、AIME2024和GPQA-Diamond上的实验表明，语义命题检索在结构化领域提升准确率，但在异构领域暴露失败模式。对推理轨迹的几何分析显示，缓存检索在嵌入空间中引发方向性偏置，形成一致改善（提升基线准确率）和破坏性（降低基线准确率）吸引子。结果揭示了基于相似性的记忆在自我改进型大模型推理中的优势与局限。
原文摘要 · Abstract (English)
Iterative summarization based reasoning frameworks such as InftyThink enable long-horizon reasoning in large language models (LLMs) by controlling context growth, but they repeatedly regenerate similar reasoning strategies across tasks. We introduce InftyThink with Cross-Chain Memory, an extension that augments iterative reasoning with an embedding-based semantic cache of previously successful reasoning patterns. At each reasoning step, the model retrieves and conditions on the most semantically similar stored lemmas, guiding inference without expanding the context window indiscriminately. Experiments on MATH500, AIME2024, and GPQA-Diamond demonstrate that semantic lemma retrieval improves accuracy in structured domains while exposing failure modes in tests that include heterogeneous domains. Geometric analyses of reasoning trajectories reveal that cache retrieval induces directional biases in embedding space, leading to consistent fix (improve baseline accuracy) and break (degradation in baseline accuracy) attractors. Our results highlight both the benefits and limits of similarity-based memory for self-improving LLM reasoning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。